You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@commons.apache.org by Morgan Delagrange <md...@yahoo.com> on 2003/01/03 17:26:03 UTC
[jelly] gump descriptor needs update
Hey all,
Jelly failed in GUMP because werkz did not build.
Since Jelly doesn't depend on werkz anymore, I should
probably update the descriptor. If somebody could
give me karma to GUMP and point me to the right file,
I'll take care of it. I see a gump.xml in the Jelly
tree, but I don't think it's actually used; the nag
email address is wrong.
- Morgan
=====
Morgan Delagrange
http://jakarta.apache.org/taglibs
http://jakarta.apache.org/commons
http://axion.tigris.org
http://jakarta.apache.org/watchdog
__________________________________________________
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
http://mailplus.yahoo.com
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [jelly] gump descriptor needs update
Posted by James Strachan <ja...@yahoo.co.uk>.
From: "Morgan Delagrange" <md...@yahoo.com>
> Hey all,
>
> Jelly failed in GUMP because werkz did not build.
> Since Jelly doesn't depend on werkz anymore, I should
> probably update the descriptor. If somebody could
> give me karma to GUMP and point me to the right file,
> I'll take care of it. I see a gump.xml in the Jelly
> tree, but I don't think it's actually used; the nag
> email address is wrong.
We could try regenerating the GUMP descriptor and try to use that? (It'd be
nice to try get Maven's generation of gump descriptors working).
James
-------
http://radio.weblogs.com/0112098/
__________________________________________________
Do You Yahoo!?
Everything you'll ever need on one web page
from News and Sport to Email and Music Charts
http://uk.my.yahoo.com
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [jelly] gump descriptor needs update
Posted by James Strachan <ja...@yahoo.co.uk>.
From: "Sam Ruby" <ru...@apache.org>
> Morgan Delagrange wrote:
> > WRT. the unit test failure, I'm still trying to track
> > it down. "It works on my machine." ;0 It's odd that
> > you are seeing one test executed. If I'm
> > understanding the Maven output correctly, you should
> > be seeing 12 results. It may just be a difference in
> > the way the test output is handled.
>
> I've verified that the unit test passes with the latest bootstrapped
> maven. Now I am going to try one by one substituting the jars provided
> by maven with the ones produced by gump to narrow down the failure.
I'm wondering if this test fails due to CVS HEAD of Ant. (It might be
related to property handling changes Costin mailed about recently).
James
-------
http://radio.weblogs.com/0112098/
__________________________________________________
Do You Yahoo!?
Everything you'll ever need on one web page
from News and Sport to Email and Music Charts
http://uk.my.yahoo.com
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [xerces][jelly] html test failure
Posted by Sam Ruby <ru...@apache.org>.
dion@multitask.com.au wrote:
>
>>3) Would it be possible for the Jelly and Xerces developers to work
>>together to debug this? For example, if the Jelly developers could
>>provide a minimal test case, that would be most helpful.
>
> Sure. The given tests pass for me in Maven, so I'm wondering if you could
> detail your environment?
If you build using the jars provided at http://gump.covalent.net/jars/
instead of the ones that that Maven downloads, you should see the
failure. In particular, I believe I have isolated it a particular jar.
More specifically, any version of xercesImpl.jar that is built after
the following change was committed:
http://marc.theaimsgroup.com/?l=xerces-cvs&m=104000199505533&w=2
- Sam Ruby
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [xerces][jelly] html test failure
Posted by Sam Ruby <ru...@apache.org>.
dion@multitask.com.au wrote:
>
>>3) Would it be possible for the Jelly and Xerces developers to work
>>together to debug this? For example, if the Jelly developers could
>>provide a minimal test case, that would be most helpful.
>
> Sure. The given tests pass for me in Maven, so I'm wondering if you could
> detail your environment?
If you build using the jars provided at http://gump.covalent.net/jars/
instead of the ones that that Maven downloads, you should see the
failure. In particular, I believe I have isolated it a particular jar.
More specifically, any version of xercesImpl.jar that is built after
the following change was committed:
http://marc.theaimsgroup.com/?l=xerces-cvs&m=104000199505533&w=2
- Sam Ruby
---------------------------------------------------------------------
To unsubscribe, e-mail: xerces-j-dev-unsubscribe@xml.apache.org
For additional commands, e-mail: xerces-j-dev-help@xml.apache.org
Re: [xerces][jelly] html test failure
Posted by Sam Ruby <ru...@apache.org>.
Craig R. McClanahan wrote:
>
> Well, I'm sorry you guys still have the problem, but I'm happy it wasn't
> my fault :-).
I wouldn't be so sure. If I roll back to before the following commit,
all works:
http://marc.theaimsgroup.com/?l=jakarta-commons-dev&m=104049921212466&w=2
If I roll forward to just include that additional commit,
jelly.ant.TestJelly fails.
Mr. McClanahan, I see your fingerprints on that commit.
- Sam Ruby
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [xerces][jelly] html test failure
Posted by "Craig R. McClanahan" <cr...@apache.org>.
On Fri, 3 Jan 2003, Sam Ruby wrote:
> Date: Fri, 03 Jan 2003 15:46:30 -0500
> From: Sam Ruby <ru...@apache.org>
> Reply-To: Jakarta Commons Developers List <co...@jakarta.apache.org>
> To: commons-dev@jakarta.apache.org
> Subject: Re: [xerces][jelly] html test failure
>
> Craig R. McClanahan wrote:
> >
> >>Until the beanutils / jelly.ant.TestJelly problem is addressed, we won't
> >>even get that far.
> >
> > It is *possible* that this problem was really due to a Beanutils bug that
> > I just checked in the fix for:
> >
> > http://nagoya.apache.org/bugzilla/show_bug.cgi?id=15773
> >
> > The next GUMP run should pick up the fixed version, so we'll soon see if
> > that was the culprit.
>
> It does not appear to have fixed the problem, or even changed the symptoms.
>
> Sorry.
Well, I'm sorry you guys still have the problem, but I'm happy it wasn't
my fault :-).
>
> - Sam Ruby
Craig
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [xerces][jelly] html test failure
Posted by Sam Ruby <ru...@apache.org>.
Craig R. McClanahan wrote:
>
>>Until the beanutils / jelly.ant.TestJelly problem is addressed, we won't
>>even get that far.
>
> It is *possible* that this problem was really due to a Beanutils bug that
> I just checked in the fix for:
>
> http://nagoya.apache.org/bugzilla/show_bug.cgi?id=15773
>
> The next GUMP run should pick up the fixed version, so we'll soon see if
> that was the culprit.
It does not appear to have fixed the problem, or even changed the symptoms.
Sorry.
- Sam Ruby
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [xerces][jelly] html test failure
Posted by Elena Litani <el...@ca.ibm.com>.
Sam Ruby wrote:
> This test fails when running with the latest (from cvs) nekohtml and a
> 2002-12-15 17:26:22 version of xerces.
There was a SAX bug in the 2002-12-15 commit that was fixed on
2002-12-19.
Can you check if Jelly still fails with the most recent Xerces +
nekohtml builds?
If yes, do you have any standalone test case that we can run to
reproduce a problem?
Thank you,
--
Elena Litani / IBM Toronto
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [xerces][jelly] html test failure
Posted by Elena Litani <el...@ca.ibm.com>.
Sam Ruby wrote:
> This test fails when running with the latest (from cvs) nekohtml and a
> 2002-12-15 17:26:22 version of xerces.
There was a SAX bug in the 2002-12-15 commit that was fixed on
2002-12-19.
Can you check if Jelly still fails with the most recent Xerces +
nekohtml builds?
If yes, do you have any standalone test case that we can run to
reproduce a problem?
Thank you,
--
Elena Litani / IBM Toronto
---------------------------------------------------------------------
To unsubscribe, e-mail: xerces-j-dev-unsubscribe@xml.apache.org
For additional commands, e-mail: xerces-j-dev-help@xml.apache.org
Re: [xerces][jelly] html test failure
Posted by "Craig R. McClanahan" <cr...@apache.org>.
> Until the beanutils / jelly.ant.TestJelly problem is addressed, we won't
> even get that far.
It is *possible* that this problem was really due to a Beanutils bug that
I just checked in the fix for:
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=15773
The next GUMP run should pick up the fixed version, so we'll soon see if
that was the culprit.
Craig
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [xerces][jelly] html test failure
Posted by di...@multitask.com.au.
Morgan Delagrange <md...@yahoo.com> wrote on 04/01/2003 07:04:02 AM:
> I'm actually all for moving ant to a separate build as
> well. James had suggested that the ant tags be part
> of the core distribution, but my inclination is to
> move any taglib other than 'core' that's not required
> in the unit tests. The only other tag I'm sure we
That makes some sense.
> need to have around is 'junit'. If the 'ant' build is
> totally contained in the package, I would just as soon
> move it. James, dIon, are you out there? :)
Yep. Sounds ok to me.
--
dIon Gillard, Multitask Consulting
Blog: http://www.freeroller.net/page/dion/Weblog
Work: http://www.multitask.com.au
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [xerces][jelly] html test failure
Posted by Morgan Delagrange <md...@yahoo.com>.
--- Sam Ruby <ru...@apache.org> wrote:
> Morgan Delagrange wrote:
> >
> > It may be a bug in nekohtml. We're pointing to
> 0.6.5,
> > look at version 0.6.7:
> >
> >
>
http://www.apache.org/~andyc/neko/doc/html/changes.html
> >
> > I'll see if I can talk the Maven folks into
> uploading
> > the most recent release. It might fix the
> problem.
> > We'd also have to point Maven to a more recent
> release
> > of Xerxes (currently we use 2.0.2) and see what
> else
> > breaks. I'm willing to do that all now.
>
> This test passes when running with the latest (from
> cvs) nekohtml and a
> 2002-12-15 17:26:20 version of xerces.
>
> This test fails when running with the latest (from
> cvs) nekohtml and a
> 2002-12-15 17:26:22 version of xerces.
>
> Draw your own conclusions.
Good point. :)
> > I moved HTML to a separate build, so you should be
> > able to get farther now. The 'util' taglib also
> has a
> > dependency on nekohtml, so I couldn't remove that
> from
> > the build. Since nekohtml is pre-1.0, that makes
> me
> > less inclined to include util in the core release.
>
> Until the beanutils / jelly.ant.TestJelly problem is
> addressed, we won't
> even get that far.
I'm actually all for moving ant to a separate build as
well. James had suggested that the ant tags be part
of the core distribution, but my inclination is to
move any taglib other than 'core' that's not required
in the unit tests. The only other tag I'm sure we
need to have around is 'junit'. If the 'ant' build is
totally contained in the package, I would just as soon
move it. James, dIon, are you out there? :)
- Morgan
> The good news is that these are the only two
> failures.
>
> - Sam Ruby
>
>
> --
> To unsubscribe, e-mail:
> <ma...@jakarta.apache.org>
> For additional commands, e-mail:
> <ma...@jakarta.apache.org>
>
=====
Morgan Delagrange
http://jakarta.apache.org/taglibs
http://jakarta.apache.org/commons
http://axion.tigris.org
http://jakarta.apache.org/watchdog
__________________________________________________
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
http://mailplus.yahoo.com
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [xerces][jelly] html test failure
Posted by Sam Ruby <ru...@apache.org>.
Morgan Delagrange wrote:
>
> It may be a bug in nekohtml. We're pointing to 0.6.5,
> look at version 0.6.7:
>
> http://www.apache.org/~andyc/neko/doc/html/changes.html
>
> I'll see if I can talk the Maven folks into uploading
> the most recent release. It might fix the problem.
> We'd also have to point Maven to a more recent release
> of Xerxes (currently we use 2.0.2) and see what else
> breaks. I'm willing to do that all now.
This test passes when running with the latest (from cvs) nekohtml and a
2002-12-15 17:26:20 version of xerces.
This test fails when running with the latest (from cvs) nekohtml and a
2002-12-15 17:26:22 version of xerces.
Draw your own conclusions.
> I moved HTML to a separate build, so you should be
> able to get farther now. The 'util' taglib also has a
> dependency on nekohtml, so I couldn't remove that from
> the build. Since nekohtml is pre-1.0, that makes me
> less inclined to include util in the core release.
Until the beanutils / jelly.ant.TestJelly problem is addressed, we won't
even get that far.
The good news is that these are the only two failures.
- Sam Ruby
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [xerces][jelly] html test failure
Posted by Sam Ruby <ru...@apache.org>.
Morgan Delagrange wrote:
>
> It may be a bug in nekohtml. We're pointing to 0.6.5,
> look at version 0.6.7:
>
> http://www.apache.org/~andyc/neko/doc/html/changes.html
>
> I'll see if I can talk the Maven folks into uploading
> the most recent release. It might fix the problem.
> We'd also have to point Maven to a more recent release
> of Xerxes (currently we use 2.0.2) and see what else
> breaks. I'm willing to do that all now.
This test passes when running with the latest (from cvs) nekohtml and a
2002-12-15 17:26:20 version of xerces.
This test fails when running with the latest (from cvs) nekohtml and a
2002-12-15 17:26:22 version of xerces.
Draw your own conclusions.
> I moved HTML to a separate build, so you should be
> able to get farther now. The 'util' taglib also has a
> dependency on nekohtml, so I couldn't remove that from
> the build. Since nekohtml is pre-1.0, that makes me
> less inclined to include util in the core release.
Until the beanutils / jelly.ant.TestJelly problem is addressed, we won't
even get that far.
The good news is that these are the only two failures.
- Sam Ruby
---------------------------------------------------------------------
To unsubscribe, e-mail: xerces-j-dev-unsubscribe@xml.apache.org
For additional commands, e-mail: xerces-j-dev-help@xml.apache.org
Re: [xerces][jelly] html test failure
Posted by di...@multitask.com.au.
Morgan Delagrange <md...@yahoo.com> wrote on 04/01/2003 06:37:20 AM:
>
> --- Sam Ruby <ru...@apache.org> wrote:
> > The following test failure:
> >
> > >
> >
>
http://cvs.apache.org/~rubys/TEST-org.apache.commons.jelly.html.TestJelly.txt
> > >
> >
>
http://cvs.apache.org/~rubys/TEST-org.apache.commons.jelly.html.TestJelly.xml
> >
> > Was caused by the following commit to Xerces:
> >
> > >
> >
> http://marc.theaimsgroup.com/?l=xerces-cvs&m=104000199505533&w=2
> >
> > The questions I would like to ask to the Jelly
> > developers is:
> >
> > 1) If this turns out to be a bug in Xerces, would
> > you prefer that it
> > gets fixed before or after the next Xerces release?
> >
> > 2) If this turns out to be a bug in Jelly, would you
> > prefer that it gets
> > fixed before or after the first Jelly release?
>
> It may be a bug in nekohtml. We're pointing to 0.6.5,
> look at version 0.6.7:
>
>
> http://www.apache.org/~andyc/neko/doc/html/changes.html
>
> I'll see if I can talk the Maven folks into uploading
> the most recent release. It might fix the problem.
> We'd also have to point Maven to a more recent release
> of Xerxes (currently we use 2.0.2) and see what else
> breaks. I'm willing to do that all now.
0.7.1 of nekohtml is now on ibiblio.
--
dIon Gillard, Multitask Consulting
Blog: http://www.freeroller.net/page/dion/Weblog
Work: http://www.multitask.com.au
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [xerces][jelly] html test failure
Posted by Morgan Delagrange <md...@yahoo.com>.
--- Sam Ruby <ru...@apache.org> wrote:
> The following test failure:
>
> >
>
http://cvs.apache.org/~rubys/TEST-org.apache.commons.jelly.html.TestJelly.txt
> >
>
http://cvs.apache.org/~rubys/TEST-org.apache.commons.jelly.html.TestJelly.xml
>
> Was caused by the following commit to Xerces:
>
> >
>
http://marc.theaimsgroup.com/?l=xerces-cvs&m=104000199505533&w=2
>
> The questions I would like to ask to the Jelly
> developers is:
>
> 1) If this turns out to be a bug in Xerces, would
> you prefer that it
> gets fixed before or after the next Xerces release?
>
> 2) If this turns out to be a bug in Jelly, would you
> prefer that it gets
> fixed before or after the first Jelly release?
It may be a bug in nekohtml. We're pointing to 0.6.5,
look at version 0.6.7:
http://www.apache.org/~andyc/neko/doc/html/changes.html
I'll see if I can talk the Maven folks into uploading
the most recent release. It might fix the problem.
We'd also have to point Maven to a more recent release
of Xerxes (currently we use 2.0.2) and see what else
breaks. I'm willing to do that all now.
I moved HTML to a separate build, so you should be
able to get farther now. The 'util' taglib also has a
dependency on nekohtml, so I couldn't remove that from
the build. Since nekohtml is pre-1.0, that makes me
less inclined to include util in the core release.
> And most importantly:
>
> 3) Would it be possible for the Jelly and Xerces
> developers to work
> together to debug this? For example, if the Jelly
> developers could
> provide a minimal test case, that would be most
> helpful.
Hopefully it won't come to that.
- Morgan
> Thanks.
>
> - Sam Ruby
>
>
> --
> To unsubscribe, e-mail:
> <ma...@jakarta.apache.org>
> For additional commands, e-mail:
> <ma...@jakarta.apache.org>
>
=====
Morgan Delagrange
http://jakarta.apache.org/taglibs
http://jakarta.apache.org/commons
http://axion.tigris.org
http://jakarta.apache.org/watchdog
__________________________________________________
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
http://mailplus.yahoo.com
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [xerces][jelly] html test failure
Posted by di...@multitask.com.au.
Sam Ruby <ru...@apache.org> wrote on 04/01/2003 06:12:44 AM:
> The following test failure:
>
> > http://cvs.apache.org/~rubys/TEST-org.apache.commons.jelly.html.
> TestJelly.txt
> > http://cvs.apache.org/~rubys/TEST-org.apache.commons.jelly.html.
> TestJelly.xml
>
> Was caused by the following commit to Xerces:
>
> > http://marc.theaimsgroup.com/?l=xerces-cvs&m=104000199505533&w=2
>
> The questions I would like to ask to the Jelly developers is:
>
> 1) If this turns out to be a bug in Xerces, would you prefer that it
> gets fixed before or after the next Xerces release?
Before.
> 2) If this turns out to be a bug in Jelly, would you prefer that it gets
> fixed before or after the first Jelly release?
Before.
> And most importantly:
>
> 3) Would it be possible for the Jelly and Xerces developers to work
> together to debug this? For example, if the Jelly developers could
> provide a minimal test case, that would be most helpful.
Sure. The given tests pass for me in Maven, so I'm wondering if you could
detail your environment?
I'm still trying to get Gump's gen and build performing acceptably for me.
--
dIon Gillard, Multitask Consulting
Blog: http://www.freeroller.net/page/dion/Weblog
Work: http://www.multitask.com.au
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
[xerces][jelly] html test failure
Posted by Sam Ruby <ru...@apache.org>.
The following test failure:
> http://cvs.apache.org/~rubys/TEST-org.apache.commons.jelly.html.TestJelly.txt
> http://cvs.apache.org/~rubys/TEST-org.apache.commons.jelly.html.TestJelly.xml
Was caused by the following commit to Xerces:
> http://marc.theaimsgroup.com/?l=xerces-cvs&m=104000199505533&w=2
The questions I would like to ask to the Jelly developers is:
1) If this turns out to be a bug in Xerces, would you prefer that it
gets fixed before or after the next Xerces release?
2) If this turns out to be a bug in Jelly, would you prefer that it gets
fixed before or after the first Jelly release?
And most importantly:
3) Would it be possible for the Jelly and Xerces developers to work
together to debug this? For example, if the Jelly developers could
provide a minimal test case, that would be most helpful.
Thanks.
- Sam Ruby
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
[xerces][jelly] html test failure
Posted by Sam Ruby <ru...@apache.org>.
The following test failure:
> http://cvs.apache.org/~rubys/TEST-org.apache.commons.jelly.html.TestJelly.txt
> http://cvs.apache.org/~rubys/TEST-org.apache.commons.jelly.html.TestJelly.xml
Was caused by the following commit to Xerces:
> http://marc.theaimsgroup.com/?l=xerces-cvs&m=104000199505533&w=2
The questions I would like to ask to the Jelly developers is:
1) If this turns out to be a bug in Xerces, would you prefer that it
gets fixed before or after the next Xerces release?
2) If this turns out to be a bug in Jelly, would you prefer that it gets
fixed before or after the first Jelly release?
And most importantly:
3) Would it be possible for the Jelly and Xerces developers to work
together to debug this? For example, if the Jelly developers could
provide a minimal test case, that would be most helpful.
Thanks.
- Sam Ruby
---------------------------------------------------------------------
To unsubscribe, e-mail: xerces-j-dev-unsubscribe@xml.apache.org
For additional commands, e-mail: xerces-j-dev-help@xml.apache.org
Re: [jelly] gump descriptor needs update
Posted by Morgan Delagrange <md...@yahoo.com>.
--- Sam Ruby <ru...@apache.org> wrote:
> Morgan Delagrange wrote:
> >
> > If you'd rather defer debugging "html", I could
> move
> > it into a separate build. It would probably only
> take
> > a couple of minutes, as long as there aren't any
> > interdependencies with other tags.
>
> I've uploaded the junit output. Looks like a null
> pointer exception
> inside of Xerces.
>
>
http://cvs.apache.org/~rubys/TEST-org.apache.commons.jelly.html.TestJelly.txt
>
http://cvs.apache.org/~rubys/TEST-org.apache.commons.jelly.html.TestJelly.xml
>
> - Sam Ruby
>
I'm moving it to a separate build. We'll have plenty
of time to sort out the problems once we hook the
individual taglibs into GUMP.
- Morgan
=====
Morgan Delagrange
http://jakarta.apache.org/taglibs
http://jakarta.apache.org/commons
http://axion.tigris.org
http://jakarta.apache.org/watchdog
__________________________________________________
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
http://mailplus.yahoo.com
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [jelly] gump descriptor needs update
Posted by Sam Ruby <ru...@apache.org>.
Morgan Delagrange wrote:
>
> If you'd rather defer debugging "html", I could move
> it into a separate build. It would probably only take
> a couple of minutes, as long as there aren't any
> interdependencies with other tags.
I've uploaded the junit output. Looks like a null pointer exception
inside of Xerces.
http://cvs.apache.org/~rubys/TEST-org.apache.commons.jelly.html.TestJelly.txt
http://cvs.apache.org/~rubys/TEST-org.apache.commons.jelly.html.TestJelly.xml
- Sam Ruby
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [jelly] gump descriptor needs update
Posted by Sam Ruby <ru...@apache.org>.
Sam Ruby wrote:
>
>> I've verified that the unit test passes with the latest bootstrapped
>> maven. Now I am going to try one by one substituting the jars
>> provided by maven with the ones produced by gump to narrow down the
>> failure.
>
> That test passes if beanutils 1.4.1 is substituted for the latest
> version from cvs. Will try to narrow down further.
I've narrowed this down to a single commit, consisting of two lines:
http://marc.theaimsgroup.com/?l=jakarta-commons-dev&m=104049921212466&w=2
- Sam Ruby
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [jelly] gump descriptor needs update
Posted by Morgan Delagrange <md...@yahoo.com>.
--- Sam Ruby <ru...@apache.org> wrote:
> Sam Ruby wrote:
> >
> > I've verified that the unit test passes with the
> latest bootstrapped
> > maven. Now I am going to try one by one
> substituting the jars provided
> > by maven with the ones produced by gump to narrow
> down the failure.
>
> That test passes if beanutils 1.4.1 is substituted
> for the latest
> version from cvs. Will try to narrow down further.
>
> The next failure is html, which of course will now
> need to be built
> separately. The way I normally handle such things
> is to only build the
> jar, go on to build the other things that may be
> necessary, and then
> come back and build the test target, unfortunately
> with the current
> build.xml, that is not an option as jar depends on
> test.
>
> > - Sam Ruby
>
If you'd rather defer debugging "html", I could move
it into a separate build. It would probably only take
a couple of minutes, as long as there aren't any
interdependencies with other tags.
- Morgan
=====
Morgan Delagrange
http://jakarta.apache.org/taglibs
http://jakarta.apache.org/commons
http://axion.tigris.org
http://jakarta.apache.org/watchdog
__________________________________________________
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
http://mailplus.yahoo.com
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [jelly] gump descriptor needs update
Posted by Sam Ruby <ru...@apache.org>.
Sam Ruby wrote:
>
> I've verified that the unit test passes with the latest bootstrapped
> maven. Now I am going to try one by one substituting the jars provided
> by maven with the ones produced by gump to narrow down the failure.
That test passes if beanutils 1.4.1 is substituted for the latest
version from cvs. Will try to narrow down further.
The next failure is html, which of course will now need to be built
separately. The way I normally handle such things is to only build the
jar, go on to build the other things that may be necessary, and then
come back and build the test target, unfortunately with the current
build.xml, that is not an option as jar depends on test.
> - Sam Ruby
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [jelly] gump descriptor needs update
Posted by Morgan Delagrange <md...@yahoo.com>.
--- Sam Ruby <ru...@apache.org> wrote:
> Morgan Delagrange wrote:
> >
> > Interesting, you removed some dependencies that
> are
> > still in the Jelly script. I'll have to try those
> on
> > my own, see if we can nix them.
>
> FYI: I actually removed them all, then added them
> back in one by one
> until the compile worked.
I'm not that thorough. On a 500mhz Thinkpad with 128
megs of RAM, the Maven-based Jelly build CRAWLS even
when I shut down all the non-essential stuff.
I found that I could take hsqldb out of the classpath
with no problem. My guess is that the SQL tags used
to use them in the unit tests, but now they use
tag-constructed result sets. I left JDBC 2.0 in the
classpath for older JVMs. When you get to the
validate unit tests, they will fail; adding relaxng
back into the classpath should fix it.
>
> > WRT. the unit test failure, I'm still trying to
> track
> > it down. "It works on my machine." ;0 It's odd
> that
> > you are seeing one test executed. If I'm
> > understanding the Maven output correctly, you
> should
> > be seeing 12 results. It may just be a difference
> in
> > the way the test output is handled.
>
> I've verified that the unit test passes with the
> latest bootstrapped
> maven. Now I am going to try one by one
> substituting the jars provided
> by maven with the ones produced by gump to narrow
> down the failure.
You are a machine.
> - Sam Ruby
>
>
>
>
> --
> To unsubscribe, e-mail:
> <ma...@jakarta.apache.org>
> For additional commands, e-mail:
> <ma...@jakarta.apache.org>
>
=====
Morgan Delagrange
http://jakarta.apache.org/taglibs
http://jakarta.apache.org/commons
http://axion.tigris.org
http://jakarta.apache.org/watchdog
__________________________________________________
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
http://mailplus.yahoo.com
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [jelly] gump descriptor needs update
Posted by Sam Ruby <ru...@apache.org>.
Morgan Delagrange wrote:
>
> Interesting, you removed some dependencies that are
> still in the Jelly script. I'll have to try those on
> my own, see if we can nix them.
FYI: I actually removed them all, then added them back in one by one
until the compile worked.
> WRT. the unit test failure, I'm still trying to track
> it down. "It works on my machine." ;0 It's odd that
> you are seeing one test executed. If I'm
> understanding the Maven output correctly, you should
> be seeing 12 results. It may just be a difference in
> the way the test output is handled.
I've verified that the unit test passes with the latest bootstrapped
maven. Now I am going to try one by one substituting the jars provided
by maven with the ones produced by gump to narrow down the failure.
- Sam Ruby
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [jelly] gump descriptor needs update
Posted by Morgan Delagrange <md...@yahoo.com>.
--- Sam Ruby <ru...@apache.org> wrote:
>
>
> The dependencies have already been removed:
>
>
http://cvs.apache.org/viewcvs/jakarta-gump/project/jakarta-commons-sandbox.xml.diff?r1=1.106&r2=1.107
>
Interesting, you removed some dependencies that are
still in the Jelly script. I'll have to try those on
my own, see if we can nix them.
WRT. the unit test failure, I'm still trying to track
it down. "It works on my machine." ;0 It's odd that
you are seeing one test executed. If I'm
understanding the Maven output correctly, you should
be seeing 12 results. It may just be a difference in
the way the test output is handled.
- Morgan
=====
Morgan Delagrange
http://jakarta.apache.org/taglibs
http://jakarta.apache.org/commons
http://axion.tigris.org
http://jakarta.apache.org/watchdog
__________________________________________________
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
http://mailplus.yahoo.com
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [jelly] gump descriptor needs update
Posted by Sam Ruby <ru...@apache.org>.
Morgan Delagrange wrote:
>
> Jelly failed in GUMP because werkz did not build.
> Since Jelly doesn't depend on werkz anymore, I should
> probably update the descriptor. If somebody could
> give me karma to GUMP and point me to the right file,
> I'll take care of it. I see a gump.xml in the Jelly
> tree, but I don't think it's actually used; the nag
> email address is wrong.
The dependencies have already been removed:
http://cvs.apache.org/viewcvs/jakarta-gump/project/jakarta-commons-sandbox.xml.diff?r1=1.106&r2=1.107
As far as updating the nag address goes, I think you will like this answer:
http://jakarta.apache.org/gump/faq.html#8
- Sam Ruby
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>